1. Autocommunication as a Mechanism of Negative Influence

This paper aims to interpret Jurij Lotman’s thorny concept of autocommunication in terms of the trite theme of ‘negative influence’ in comparative literature and crosscultural studies. I shall raise and attempt to answer three related questions: (1) What is‘negative influence’? (2) How ‘negative influence’ can be recast in the Jakobsonian model of speech communication? (3) How ‘negative influence’ can be reinterpreted in terms of‘autocommunication’? (4) What implications ‘autocommunication’ has in the writing of national / transnational literary historiography?

Chang (1992, 2000, 2008) appropriates Roman Jakobson’s speech communication model and expands it—albeit with a qualitative leap, to discuss Chinese scholars’receptions of (1) John Dewey’s pragmatism and (2) Wilhelm Windleband’s historiography of philosophy in the early 20th century, and (3) American New Criticism in the 1970s as instances of negative influence. The present paper attempts, among other things, to negotiate negative influence with Lotman’s model of autocommunication.

2. What Is ‘Negative Influence’?

‘Negative influence’ is a term popularised among comparative literature scholars in the 1960s and 1970s. It refers to [the concept of] the phenomenon of literary reception in which the recipient, rather than being influenced to produce works similar to the model,reacts against his/her foreign source in the manner of counter imitation. A famous example is Bertolt Brecht’s Gegenentwurf (‘counter-design’) through which the playwright distorts his models intentionally and polemically. There have been a few other terms used to refer to ‘negative influence’. For example, Paul Valéry refers to it as ‘l’influence à rebours’(Pistorius, 1966), and André Gide as ‘l’influence par protestation’ (Laurette, 1966),and the most famous is probably Harold Bloom’s ‘anxiety of influence’ (Bloom, 1975).Alternatively, in the Slavonic tradition, Viktor M. Zhirmunsky quotes Veseolovsky in describing convergent literary currents as ‘vstrechnye techeniya’ (‘courants de rencontre’)(1967, 1969). This tradition is carried over by the Slovakian D. Ďurišin who uses the term‘convergent currents’ (1984) in his configuration of interliterary process.

All the cited variations suggest the concept is founded on the metaphor of water flow. And the etymology of ‘influence’ (in-fluere) and ‘influx’ suggests the direction of information flow from a ‘source’ to a ‘flow-in’ receptacle, an archaic ‘metalingual’top-down movement, pace Lotman, and, conceptualised in typography, a left to right horizontal movement. To rebuke Bloom’s metaphor of water flow, Thomas McFarland(1984) reinstates the meaning of astrology in that celestial bodies affect human beings,which is nonetheless still registered in a top-down movement. Furthermore, the metaphor of ‘current’ or ‘convergent currents’ suggests the possibility of two-way flow and is congenial to the Lotmanian dialectics of autocommunication and dialogisation.

农田复种指数扩大，绿肥种植逐年缩小甚至淘汰，普遍实行大宗作物长年连作，麦田禾本科杂草、阔叶杂草共生，防除难度加大。随着绿麦隆推广应用，麦田看麦娘密度迅速下降，日本看麦娘、硬草密度急剧上升，成为优势种群，伴生早熟禾、野燕麦，局部区域如东台还有毒麦。阔叶杂草稻麦轮作区以猪秧秧、大巢菜、牛繁缕等为主；南部棉旱粮地区麦田阔叶杂草以卷耳为主，伴生猪秧秧、婆婆纳、牛繁缕等；北部淮北沙土麦田播娘蒿发生严重。但80年代末至90年代初，甲、绿磺隆等超高效除草剂推广后，卷耳密度迅速下降，田间阔叶杂草以婆婆纳、猪秧秧为主，伴生卷耳、牛繁缕等，渠北旱茬麦田麦家公发展也十分迅速。

Scholars have identified two types of negative influence. First, negative influence occurs when the Addressee ‘misreads’ the Addresser’s Mes sage. Granting that all reading is misreading, one can make a provisional distinction here between ‘intentional’ and‘unintentional’ misreadings. The intentional misreading is done with a polemic thrust by the Addressee as post-processor in order to negate the influence of his model (text 1);the result (text 2) of his transcoding often appears as a parody or the afore-mentioned Brechtian counter-design of the original. The unintentional misreading takes place when the Addressee, often a post-processing mediator, lacks sufficient linguistic and literary competence to ‘reconstruct’ the Addresser’s code.

朱大可先生把大洪水的世界性叙事作为非洲神话传播的结果。如果大洪水是世界性的，那么，身处世界各地的人们也会有对大洪水的记忆，不需要延续非洲神话不可。同样面对大洪水，中国人的态度不像诺亚那样是逃避，而是治水，一次不成两次，大禹的父亲鲧没做到，大禹做到了。

Under such circumstances, the mediator’s (addressee 1 = addresser 2) transcoded product (e.g., translations as text 2) would engender for addressee 2 a mirage of text l.This happens to be the case of mainland China’s reception of John Dewey in the 1920s and Taiwan’s reception of American New Criticism in the late 1960s.

The second definition of negative influence involving two literary systems is more relevant to our discussion. Addressee 1 (agent or mediator) introduces into his own culture a foreign model (text 1), often adulterated through transcoding (text 2), as a challenge to his immediate predecessors or conser vatives in his own literary tradition. A triangle down below is thus formed. In this instance, what addressee 1 rebels against is his own literary heritage, rivalry or the dominating ‘normal science’ rather than his transcoded model (text 1). A notable example took place in Taiwan in the early 1970s, where the traditionalists(addressee 2) and the mediators (addressee 1 = agent = addresser 2) of the New Critical method staged a keen fight in the academy. The relationship can be shown in Figure 1.

Figure 1

3. A Rehearsal of Jakobson’s Communication Model

As Lotman’s autocommunication has been regarded as a revision and modification of Jakobson’s model of communication, a brief rehearsal of Jakobson (1960) and his critics will be in order.

Figure 2. Jakobson’s communication model

Figure 3. Chang’s expansion of Jakobson’s model

According to Eco (1976), the Addressee’s interpretation is affected by several‘aberrant’ factors, including his private codes and ideolog ical biases. Besides, the ‘real patrimony’ of the Addressee’s knowledge is not the knowledge he supposedly shares with the Sender. Therefore, the decoded text (message-content) is not expected to coincide with the encoded text (message-expression). This is especially true in cross-cultural communication. As Eco puts it, ‘Sometimes the addressee’s entire system of cultural units(as well as the concrete circumstances in which he lives) legitimate[s] an interpre tation that the sender would have never foreseen’ (p. 141). Other times, the Sender’s message is received as noise which produces a new culture (p. 142).

用SPSS22.0统计学软件分析本文研究中获得的数据，用卡方检验[n（%）]计数资料差异；用t检验（±s）计量资料差异。最后以P＜0.05表示统计学意义。

Where addressee 1 is identical to addresser 2 in post-processing, and there is interlingual transcoding from code 1 to code 2, e.g., from English to Chinese.

However, Lotman’s revision requires scrutiny preferably in terms of and in relation to Jakobson. A few observations can be made here. Firstly, Addresser and Addressee are not real entities, but are functions which can receive nominal and pronominal, semantic and ideological investments, including the rare case of the split personality ‘I’ – ‘I’. Secondly,substituting the pronominal ‘I’ and ‘I’ for Addresser and Addressee does not necessarily constitute self-communication. All the six functions in Jakobson are abstractions and should be understood as such in the theoretical and irreducible sense. The fallacy of pluralisation of abstractions is seen not only in the pronominal ‘I’ but also in ‘Code’ and‘Message’. Thirdly, Jakobson does not use Text, but we may take it to mean Message +Code or ‘Encoded Message’. In actual performance, a text may be framed in one code, or several, when the code is composite, multiple, hybridised but dividable into a number of(a hierarchy of) subcodes. Code as a function in the model is not to be confused with this or that code as a specific linguistic entity. Likewise, a message can be doubly and trebly encoded, but it remains once and for all coded.

That the Addresser and the Addressee would encode and decode differ ently the same ‘text’—a better word would be Schmidt’s KOMMUNIKATE (1980, p. 532)—is due to the different modes of action organised in their respective LITERATURE-systems. As Schmidt puts it, ‘Any recipient con structs his own KOMMUNIKAT related to a given TEXT in a receptional situation’ (1980, p. 534). This testifies to what Schmidt describes as literature’s polyvalence convention (1980, p. 543; 1983, p. 28). By virtue of this convention, the Addresser is deprived of his control over his KOMMUNIKAT so that it can be prevented from being appropriated first as a TEXT and then as another KOMMUNIKAT, and the Addressee is able to enjoy the freedom of post- processing his own KOMMUNIKAT.

Now what have Eco and Schmidt in common? Both Eco and Schmidt—two randomly picked up samples—focus on the Addressee’s decoding process, testifying to the change of critical climate from the text to the role of the reader in the 1970s.

As seen in Fig. 3, ‘negative influence’ can be assessed only in terms of addresser 2’s transcoding of message 1 into message 2; thus the relationship is intertextual rather than interpersonal or psychical on the part of addresser 2, let alone that of addresser 1.

4. Lotman’s Revision of Jakobson

Lotman departs from Jakobson and comments on his axis: ‘The model of ideal understanding [between addresser and addressee] does not even apply to man’s internal communication with himself since the latter situation implies the transfer of an intense dialogue within a single personality.’ He has provided a revised model to illustrate this kind of international communication as follows.

Figure 4. Lotman’s internal communication model (Lotman, 1990, 2001, Shukman’s translation)

膜分离法是根据VOCs中各组分分子大小不同，利用它们通过膜的传递速率及扩散能力的差异实现分离的工艺[16]，具有流程简单、能耗小、运行费用和设备占地面积小的优势，但需要在高压操作条件下进行，目前气体膜分离材料还处在不断发展的阶段。

6) Communication which resides in the nature of language itself: the tension between semantics and syntax [syntagm in Lotman] in the process of signification.

上述分析表明，四大自贸试验区在“一带一路”建设中的定位有所不同，因此，各个自贸试验区应根据自身优势及战略定位，差异化参与和推动“一带一路”建设。

5. Types of Lotmanian Autocommunication

To clarify Lotman’s use of autocommunication, I have initially identified six types and shall discuss them in their order of appearance. The six types are listed below.

1) Monologue as ‘internal’ communication;

2) Communication which exhibits the tension of ‘I’ as a pronoun and ‘I’ as a proper name;

This last critique in terms of metalingual function serves to explain, if explain away, Lotman’s communication of dual-structure, which manifests in three types. First,communication which combines an explicit ‘message’ [not in the Jakobsonian sense] and a syntagm deprived of semantic features amounts to Jakobson’s ‘poetry of grammar’. It’s a problem of Code which serves the metalingual function. Second, communication with an explicit vocal code and a telegraphic, cryptic written code can only be a problem of Code and also serves the metalingual function. And third, communication which resides in the nature of language itself: the tension between semantics and syntax [syntagm in Lotman]in the process of signification is a coding problem par excellence, cf. Jakobson’s two poles of language, the metaphorical and the metonymical, from Saussure’s rapports associatifs and rapports syntagmatiques (Saussure, 1968, pp. 170-175).

4) Communication of dual-structure which combines an explicit message [that transmits]information and a syntagm deprived of semantic features but reflective of the hidden consciousness [unconscious?];

Jakobson’s scheme has been constantly modified, for critics assert that it either fails to reflect the reality of communication (Eco, 1976, 1979) or fails to account for the socio-economic dimensions of ‘contact’ and the institutional force of ‘context’, both being instrumental in making successful communica tion possible (Even-Zohar, 1990).Even though Jakobson stresses that the model is dynamic and the six elements involved are mutually interactive, and therefore the arrows are not uni-directional, critics tend to read the information processing horizontally from left to right. I shall now identify three critics who have attempted to modify Jakobson: Umberto Eco, Siegfried J. Schmidt and Lotman.

5) Communication of dual-structure with an explicit vocal code and a telegraphic, cryptic written code;

The same can be said of Context and Message. The fact that Context in the model performs primarily the referential function does not mean that the word ‘Context’ can be pluralised. Therefore, there cannot be ‘contextual displacement’ because whatever it means, it belongs to the function of Context. Nor does Message transmit purely ‘referential’information only; any additional layers of information, ‘emotive’, ‘phatic’, ‘conative’,‘poetic’, ‘metalingual’ join to constitute information in toto, as has been confirmed by recent advances in Cognitive Science. Thus, the distinction between message 1 and message 2 is redundant unless a message is differently decoded as two messages by readers.

正交度本是公理设计中测量设计矩阵正交性的指标，文献[3]将其扩展到了测量DPM/MIM的正交性，作为模块之间相似性最小化的性能指标。

The simplest (and simplistic) type of autocommunication is no doubt monologue where one communicates with oneself. A beautiful statement of Louis Hjelmslev’s[unfortunately in English translation] reads: ‘[Language] is also the ultimate, indispensable sustainer of the human individual, his refuge in hours of loneliness, when the mind wrestles with existence and the conflict is resolved in the monologue of the poet and the thinker’ (Hjelmslev, 1970, p. 3). When introducing autocommunication, Lotman cites from Goethe’s Faust: ‘Two souls, alas! reside within my breast, / And each withdraws from, and repels, its brother’ (Lotman, 2004, pp. 4-5). But when two souls are engaged, are they not using ‘dialogue’ rather than ‘monologue’, as evidenced by the popular medieval genre of dialogue of body and soul? Here Lotman is dealing with an entirely different issue from Jakobson’s. It may touch upon the Addresser-oriented emotive and phatic functions and the Addressee-oriented conative function of language. More probably, it addresses the Addresser’s ‘double-articulation’ of ‘poetic’ language in a rhetorical (denotative versus connotative) and psychoanalytical (manifest versus latent) sense.

Jakobson’s model, which presupposes two actants performed, in principle, by two acteurs, may apply to the Addresser’s autocommunication only when the reified Addresser(as acteur) has undergone a personality split, i.e., one acteur assuming the functions of two actants, performing respectively on the manifest code and the latent code. Solipsism put aside, monologue, however, is always already socialised; it is part of sociolect,rather than an idiolect which Jakobson treats as a ‘perversion’. Apart from monologue,an obvious instance of autocommunication can be an author’s constantly revised and recanted autobiographical writings, as in the case of St. Augustine and William Butler Yeats.

The second type of autocommunication exhibits the tension of ‘I’ as a pronoun and ‘I’as a proper name. Lotman cites Rousseau as saying: ‘The [first] “I” in Emile is a pronoun designating he who expresses the essence of discourse in the first person.’ This ‘I’ can be transferred to others and pluralised [into a ‘we’] through the shifters (cf. Jakobson). But the ‘I’ in Confessions amounts to a proper name [like ‘Jean-Jacques’] in that it is unique:‘I am not made like any one I have seen; I dare believe I am not made like any one existing’ (2004). Lotman further observes that ‘The structure of the ‘I’ is one of the basic indices of culture’ (ibid.) and sees this aporia of ‘I’ has given rise to a new type of artistic consciousness. That the first-person narrator undergoes, as it were, a personality split into the historical personage (‘Jean-Jacques Rousseau’) and a fictional persona (‘Emile’),or even a voice evoking and nominating ‘I’ (in Saussure’s words, ‘le sujet parlant’) is a modernist commonplace. An enquiry into the narrative voice or point-of-view has to be distinguished from the semantics of proper noun in the tradition of F. L. Gottlob Frege, P.F. Strawson and Saul Kripke. In fact, as an indexical sign, ‘I’ does not ‘represent’ a proper noun even though it ‘evokes’ a speaking subject and, for that matter, an infinite number of shifting speaking subjects (‘une masse parlante’), thus occasioning la parole (Saussure)and le discours (Émile Benveniste). It has nothing to do with ‘naming’.

调查显示，企业融资贵主要原因是中间环节收费高。综合来看，企业需要支付5%～6%左右的担保费、评估费及过桥费，已接近利率水平，比浙江省平均收费高2～3个百分点。政策性担保机构业务量较少，而商业性担保机构担保费用基本超过2.5%；同时，企业反映，在林权评估领域，超过95%的项目由非专业的评估公司开展，且一般按林权评估价值额的3%～6%收取费用，评估费过高。

In ‘I-I’ communication, the second ‘I’ is distanced as an ‘other’, a third party (‘he’ /‘she’) or a ‘you’ because of the temporal-spatial disjunction. The argument that ‘in “I-s/he” system information is transferred in space, in the “I-I” system it is transferred in time’does not hold true (Lotman, 2001, p. 21), both systems being characterised by time and space disjunction. ‘I’ is a deictics, evoking the speaking subject; it’s not even a signifier that closes on a signified. The famous quote from Rimbaud ‘Je est un autre’ points out the arbitrariness of the pronoun, not renders the evoked subject into an otherwise substantial other, as in ‘Je suis un autre’. What makes sense in ‘I-I’ communication is the selfreflexivity of certain statements, oriented to the Code, thus serving the ‘metalingual’ (plus poetic) function identified by Jakobson.

3) ‘I-I’ communication where the second ‘I’ is distanced as an ‘other’, a third party [s/he] or a‘you’ because of the temporal-spatial disjunction;

6. What Can Lotman Teach Us About Negative Influence?—Dialogue Reconsidered

Lotman’s reading of Jakobson bears witness to the force of ‘negative influence’. One theoretical implication of my case studies is the production of new knowledge when cultures clash. This is evidenced by the triangular relationships among domestic establishments, revolutionaries, and imported foreign models, where two kinds of ‘negative influence’ are in operation. Whilst the interlocutors are engaged in heated debate, the third party from abroad invariably serves as a catalyst for ‘explosion’. For this, we should once again turn to Lotman on dialogue rather than autocommunication.

One of the theoretical implications of Lotman’s semiosphere is its significance in cross-cultural studies. Among the cultural mechanisms which Lotman and his colleagues have identified, dialogue and translation figure prominently. With Lotman, dialogue as well as translation—in their continued process of emission and transmission of energy—can be enacted not only between historical periods of one culture, but also between intercultural and cross-cultural systems. According to Lotman, dialogue is characterised by the discreteness of language and asymmetry in communication. Where the interlocutors alternate in give-and-take, each is capable of articulating only his discrete share of discourse, perhaps only one tiny fraction at a time. The discreteness is constituted not only by moments of articulation, but also by moments of silence because when one locutor speaks and sends information, the other has to remain silent and becomes temporarily an allocutor whose job is to decode the message s/he receives.

Since natural language is by nature unstable and subject to the caprice of temporality,the information flow is often asymmetrical and perfect communication is thus impossible.As natural language is the primary modelling system, on top of which is the secondary modelling system of culture, the phenomenon of interpreting culture becomes all the more difficult. This is especially the case in cross-cultural communication because each of the two parties involved has its own definition of culture, its own boundaries of the legitimate texts that constitute culture as well as exclude the so-called non-culture. As dialogue of cultures is inevitable in a culture’s historical evolution, such dialogue serves, curiously, a special function of its own dialogue or, in Lotman’s words, autocommunication.

Lotman projects the dialogic discreteness onto the history of a culture, where the interlocutors cease to be the indigenous versus the exogenous, because both have already been fused as historical products, but are displaced by two historical moments which engage each other in dialogue, or are charged with the semiotic task of infinite process of encoding and decoding. An example is the dialogue between a turbulent, productive moment and its relatively calm and inert-looking but fully saturated counterpart. In this sense, the autocommunication of a culture which is no longer a self-sufficient entity in itself amounts to the perennial Gadamarian self-dialogue that characterizes cultural hermeneutics. This is perhaps an alternative solution to the thorny problem of crosscultural dialogue with which we are all concerned.

针对复杂网络的社团结构检测问题，将Newman的基于模块度函数Q的矩阵特征值和特征向量提取复杂网络中社团结构的算法和神经网络算法求解矩阵特征值特征向量的算法结合起来得到一种新的基于连续神经网络的社团结构检测算法，简称CNN算法，模型为：

Put in the Chinese contexts of the 1920s and 1970s, the traditionalists’ vehement reactions to introduced models force the recipients of foreign models to engage in some kind of autocommunication between two domestic camps. Only in this extended metaphorical sense and through the macroscopic perspective can autocommunication be appropriated as a model to study cross-cultural reception studies. A profound semiotician and cultural historian, Lotman continues to shed light on our disciplines of comparative literature and semiotics of culture with his insight into the possibility of intercultural dialogue.

手术结束时已是凌晨1点05分，陈正副院长将患儿抱进新生儿科进行监护治疗，并安排好一切有关事宜后，这才换了衣服回宿舍，在回宿舍的路上，陈正副院长在便利店买了两盒方便面，作为犒劳自己这惊心动魄的一天的奖励。他说：“忙碌是难免的，但作为新生儿的守护者，一定要做到忙而不乱，看到转危为安的孩子们，作为一名医者，我感到很值得。”

References

Benveniste, É. (1966). Problèmes de linguistique générale 1. Paris: Gallimard.

Bloom, H. (1975). The anxiety of influence: A theory of poetry. New York: Oxford University Press.

Chang, H.-l. (1992). Reflections on cross-cultural literary contact: The reception of American critical discourse in Taiwan in the 1970s. Poetics: Journal for Empirical Research on Literature, the Media and the Arts, 21(1-2), 57-71.

Chang, H.-l. (2000). Hu Shih and John Dewey: ‘Scientific method’ in the May Fourth era—China 1919 and after. Comparative Criticism, 22, 91-103.

Chang, H.-l. (2008). On the ‘birth’ of historiography of ‘philosophy’ in China. Paper presented at the Invited Session (IS02) of Reframing the Historiography of Philosophy: A Dialectic Approach. XXII World Congress of Philosophy, Seoul, Korea, 30 July – 5 August, 2008.

Ďurišin, D. (1984). Theory of literary comparatistics (J. Kocmanová, Trans.). Bratislava: Veda,Publishing House of the Slovak Academy of Sciences.

Eco, U. (1976). A theory of semiotics. Bloomington, IN: Indiana University Press.

Eco, U. (1979). The role of the reader: Explorations in the semiotics of texts. Bloomington, IN:Indiana University Press.

Even-Zohar, I. (1990). Polysystem studies. Poetics Today, 11(1), 1-268.

Hjelmslev, L. (1970). Language: An Introduction (F. J. Whitfield, Trans.). Madison: The University of Wisconsin Press.

Jakobson, R. (1960). Closing statement: Linguistics and poetics. In T. A. Sebeok (Ed.), Style in language (pp. 350-377). Cambridge, MA: The MIT Press.

Laurette, P. (1966). La notion d’influence chez Valéry et Goethe. In F. Jost (Ed.), Proceedings of the fourth congress of the International Association of Comparative Literature (Vol. 2) (pp.1048-1052). The Hague: Mouton.

Lotman, J. M. (1977). Two models of communication. In D. P. Lucid (Ed. & Trans.), Soviet semiotics: An anthology (pp. 99-101). Baltimore: Johns Hopkins University Press.

Lotman, J. M. (2001). Universe of the mind: A semiotic theory of culture. (A. Shukman, Trans.).London: I B Tauris.

Lotman, J. M. (2004). Culture and explosion (M. Grishakova, Ed.; W. Clark, Trans.). Berlin:Mouton de Gruyter.

McFarland, T. (1984). Originality and imagination. Baltimore: Johns Hopkins University Press.

Pistorius, G. (1966). Le problème d’influence selon Paul Valéry. In Proceedings of the fourth congress of the International Association of Comparative Literature (Vol. 2) (pp. 1037-1047).The Hague: Mouton.

Saussure, F. de. (1968). Cours de linguistique générale. Paris: Payot.

Schmidt, S. J. (1980). Fictionality in literary and non-literary discourse. Poetics: Journal for Empirical Research on Literature, the Media and the Arts, 9, 525-546.

Schmidt, S. J. (1983). The empirical science of literature ESL: A new paradigm. Poetics: Journal for Empirical Research on Literature, the Media and the Arts, 12, 19-34.

Zhirmunsky, V. M. (1967). On the study of comparative literature. Oxford Slavonic Papers, 13, 1-13.

Zhirmunsky, V. M. (1969). Les courants littéraires en tant que phénomènes internationaux. In

N. Banaševič (Ed.), ACTES du Ve Congrès de l’Association Internationale de Littérature Comparée, Belgrade 1967 (Vol. 1) (pp. 3-21). Amsterdam: Swets & Zeitlinger.